Please create an account to participate in the Slashdot moderation system


Forgot your password?
Slashdot Deals: Cyber Monday Sale Extended! Courses ranging from coding to project management - all eLearning deals 20% off with coupon code "CYBERMONDAY20". ×

Comment Re:Far more abundant than lithium? (Score 1) 196

The cost of raw materials tends to drop as the demand rises.

Sometimes. Usually this happens when a raw material switches from being a niche product to an industrial one, but afterward this is not a general rule.

This sounds counter intuitive but what happens is that technological advances in finding and extracting raw material generate price drops. Think shale oil.

Not quite. What happened was was that reaching the peak production of conventional petroleum raised the prices high enough that expensive oil was now profitable to produce. Shale oil is more expensive than all but a few conventional oil sources. World oil prices have declined from their initial post-peak high simply because Saudi Arabia has chosen to turn on their cheap oil taps to drive down oil prices so low that expensive shale is no longer profitable. U.S. oil production stopped growing abruptly when they did this, it hasn't really declined much (it has dipped slightly) due to the fact that those expensive wells are already sunk costs, and it is more economical to keep pumping than shut them down at this point. But some have been shut down even so.

BTW, this reinforces the point I make above in this thread about how rising prices open up new, previously untapped types of reserves. But at higher prices.

Comment Re:Far more abundant than lithium? (Score 1) 196

The issue of "proven reserves" seems almost always to be misunderstood by the random person doing a little Googling about resources. "Proven resources" are always tied to some price point. It is the amount of a resource that we know with assurance can be produced under a particular cost ceiling, usually closely tied to current market prices.

This figure is in in no way a measure of how much is available for future use on Earth, the implication that people almost always make. It is an extremely conservative figure for the amount that is available for supply at current prices. The effect of full exploiting this proven reserve is never to "run out" (the conclusion invariably drawn), it is instead to raise the price somewhat, and to lead to further exploration of reserves. The reason these reserve estimates are made is to forecast near term commercial activity, not long term potential.

It is a very common pattern for "proven reserves" to expand extremely fast with even modest price increases. There is often an enormous potential reserve that sets a ceiling on how prices might rise on a resource, since once that price point is reached "suddenly" there is an immense proven reserve available. For lithium (and some other elements, like uranium) this huge reserve is sea water. Modest increases of lithium (or uranium) prices makes mining the oceans, a supply good for thousands of years (or longer) practical.

For example - that 14 million ton reserve figure? Well, the USGS is currently estimating "identified lithium resources" at 40 million tons, almost tripling the available amount they record with hardly any actual exploration involved, it is simply a matter of classification of what we already know. And most resources that are relatively abundant are also not intensively prospected for (yet) since there is no need.

Comment Re:The guy aint no Sagan... (Score 2) 372

Not true, asteroids could be brought near earth by incredibly cheap process that merely would be time consuming. A little nudge in the right place at the right time

Bringing asteroids "near" Earth (that is in an Earth crossing orbit, nearly intersecting the Earth) can be accomplished with a little nudge for some small set of asteroids - but this will NOT bring the asteroid "to Earth" (as the OP said) unless you are planning on causing a cosmic bombardment that will vaporize the asteroid in a spectacular explosion (good luck getting the people of Earth to buy into that plan).

Asteroid mining requires bringing the asteroid into an Earth orbit, which involved a very large change in velocity, not a "little nudge" and won't be cheap no matter how it is arranged.

Comment Re:Economic incentives and existential threats (Score 1) 372

We went to the moon to beat the Russians during the Cold War - an existential threat

Really? How did putting people on the Moon help us defend against the USSR?

The Cold War was not primarily a military conflict, but a socio/economic/political one. In 1965 about half of the world's population lived under Communist governments, and much of the rest of the world (by population) were either allies or friendly with one or more of the Communist powers (example: India). It was quite explicitly a confrontation between sociopolitical systems - the motivation behind the whole space race on both sides was to show which system was superior.

The moon race was an explicit calculation by the U.S. to show the superiority of U.S. society and the U.S. system. There were no bones made about this, it was part of the declared purpose.

And it succeeded in exactly the way envisioned. Not just succeeded as in "we got to the Moon", but showing that the Soviets could not compete. The Soviets actually tried to mount a rival Moon mission (two different programs in fact - a manned orbit of the Moon and a landing) but they fell far behind Apollo, and were abandoned in the early 1970s when the U.S. had already secured all the laurels for being first.

Comment Re:Cost of access is key. (Score 1) 372


So for the foreseeable future, private space travel will never happen except as an amusement for the hyper-rich, due to the amount of energy required...

A very common, completely incorrect notion. The cost of space flight has nothing to do with the amount of energy required. If rocket fuel were completely free it would not significantly affect the cost of space flight; the fuel cost on a rocket launch is roughly 0.01% of the overall cost. The high cost is entirely due to the cost of space flight hardware, which costs $10,000/lb or more, and is currently non-reusable, and the cost of the ground infrastructure required to support it.

Recall that even a mundane piece of aerospace hardware, like a Boeing Dreamliner, costs more than $1,000/lb. If you could only use it once, airline travel would be only for billionaires.

Comment Re: Scrum Was Never Alive (Score 1) 371

nd if it can't be broken down into something that can be completed by development, tested by QA, and verified by stakeholders in a single sprint? In that case, Agile does not allow your software project to be completed. Using Agile is Russian Roulette.

What nonsense. First, you're confusing Agile (a broad concept of software development) with scrum (a specific methodology that uses sprints). Second, anything can be broken down into bite-sized pieces that can be coded and tested in a sprint. "Verfiied by a stakeholder" is a very broad idea, unless you're writing simple CRUD software with a GUI where is has a simple, specific meaning. Writing complex back-end infrastructure code? You just need whatever senior engineer is guiding the process involved in your CR, which you're probably doing anyhow.

The important thing is to integrate early and often, test as you go, and not call a task done when there's lingering work.

Too true - but in the AC's partial defense, there are a legion of Agile methodology priests who issue decrees about how this and such must be done by definition (check out some comments on this thread), and any deviation commits the sin of not practicing True Scrum (or whatever). So the AC may be excused for taking such zealotry to heart and thinking its was anything more than bad advice from self-important blow-hards (even if they may be well know and well paid self-important blow-hards).

Comment Re:Scrum Was Never Alive (Score 2) 371

Any big architecture problem that takes more than one Sprint cannot be done with Agile. ...

Agile cannot produce good software since by definition you can't do anything that takes longer than a Sprint.

This is untrue. You create epics, which capture your big architectural issues, which lead to a series of stories for implementation over many sprints.

(My caveat about all SW methodologies - any devotion to a doctrine is a heresy. Methodologies are but tools to help organize work, and must be customized to match the environment in which is used. Organizations vary with size, requirements dynamism, team skill level, quality of management support, etc., etc. and no list of manifesto decrees can properly address them in all circumstances.)

Comment Re:Does this work for textbooks? (Score 1) 221

Your concerns are valid. No, OCR does not handle mathematical formulas very well (in any that I have seen). But remember - OCR does not replace the image, it only augments it (at least if your doing PDFs, I don't know about eBook formats). You are still reading the scanned image. OCR simply provides fast searching and indexing capabilities, a huge win. So formula's can be searched for? Well, in my experience, no math or science book include formulasin its indices anyway so this is no different (the names of formulas yes, the formulas themselves, no).

Comment I Am Giving It A Try (Score 1) 221

Lots, and lots and lots of reasons to dis this offering here. No new tech, just buy eBooks at $10 a pop, who wants hundreds of books on a device, what's so hard about destroying a book, OSS software already exists that does this, etc. etc.

Here's the thing for me: I want a research library I can take where ever I go. I am a heavy research library book user, and I buy a lot of used books, trying to get out of print texts. When I need a book, I need that exact book, and no substitute will do, because none exists. No, many of the books I want CANNOT be downloaded because someone else scanned it, or as an eBook. I cannot destroy library books, and if I own the book I'd rather have the paper copy too. Whether something is possible with similar tech and software is irrelevant. It needs to be fast and convenient. A well integrated system to this is worth a lot, a lot more than $234 (the final cost if you buy it now). I have worked with a number of Windows and Linux-based processing chains for scanning, page clean-up, compression, OCR, etc., and they are painful to use without exception, some are more painful than others, but none is anything like painless.

Will this really do the job I hope it will? I don't know, but I will find out.

Comment Re:This is why.. (Score 1) 345

Also, $40 billion on R&D vs roughly $2.5 billion in advertising by the top 10 Pharma companies in the US. Again, you are off by a ratio of 16-to-1.

If you want to cite CBO reports (a good starting point in almost all cases) then you are low-balling your advertising claim by a factor of 8-1. Here is a 2008 CBO report on drug advertising and it shows that this amounts to $20.5 billion. Even if you pretend that only direct-to-consumer advertising is real advertising, that alone is $4.5 billion, nearly double your weird low-ball claim.

Comment Use Regularly Scheduled Leap Seconds (Score 1) 291

If you don't want the solar time and UTC to drift many seconds out of synch, but are willing to let the alignment between the two to drift by more than a second at any given time, then using the long-term average rate of rotation slowing to calculate regularly scheduled leap seconds seems the way to go.

We add whole leap days at regular intervals, but it does not seem to cause any problems because everyone knows when they occur, even millenia in advance. It is not possible to schedule leap seconds that far in advance, but using an average trend over the last 20 years (perhaps), and the cumulative planning miss over the last decade, to plan for the next decade (say) would not allow deviations of more than a couple of seconds from solar time to ever accumulate.

Look at a chart of the deceleration of rotation over decades.

Comment Re:SO when you pay people... (Score 2) 500

There are "No True Scotsmen" are there?

The claim is made that anything the least bit "socialist" must fail since everything socialist always fails. But when it is pointed out that there are numerous nations far more socialist than the U.S., with many, many "socialist" policies in practice - which are succeeding very well, then the claim is made that "they don't count 'cuz they ain't completely, totally socialist"!

1000 pains = 1 Megahertz